0 comments

Technical Overview of Microsoft Forefront Identity Manager 2010 R2

Published on Sunday, May 22, 2011 in ,

On Henrik Nilson’s blog I found a link to a session from Tech-Ed: Technical Overview of Microsoft Forefront Identity Manager 2010 R2 It’s definitely worth watching. If you’re in a hurry though, here’s a brief summary:

  • Support for Extranet Password Reset (other browsers/platforms, not domain-joined)
  • Reporting: group membership and object change history
  • Extensible MA framework
  • Performance improvements
  • Best Practices Analyzer
  • Add-in Support for Outlook 2010 x86 & x64
  • Support for SharePoint 2010

Password Reset

image

Very neat: support for password resets from PC’s which are not joined to the domain! So this means the extranet password reset scenario is added to FIM 2010 R2! And because no fancy ActiveX is used this should work just fine from other browsers/platforms as well.

image

This extranet scenario is made available through a new ASP.net password registration/reset portal. This portal is not based upon Sharepoint. They also added the option to add an additional Q/A gate for people resetting their password externally. You could see these as coming from an untrusted location and requiring to answer more questions than someone doing it from on your corporate LAN.

Reporting

image

This functionality will be made available through a custom FIM MP for SCSM. The SCSM data warehouse is a requirement. If you don’t have one, you’ll be allowed to use one without additional licensing costs.

image

Extensible MA Framework

image

Outlook 2010 & Sharepoint 2010 support

image

12 comments

Running PowerShell Scripts From An UNC Path (Share)

Published on Sunday, May 15, 2011 in ,

Introduction

Lately a colleague of mine was struggling with the following: he wanted a script to be ran from within a Startup Script GPO. Now the problem he was encountering was the following Security Warning from PowerShell:

image

In words:

Security Warning
Run only scripts that you trust. While scripts from the Internet can be useful, this script can potentially harm your
computer. Do you want to run \\file.setspn.com\scripts\script.ps1?
[D] Do not run  [R] Run once  [S] Suspend  [?] Help (default is "D"):

Now if you have to run something like this occasionally, you can type R and get on it with it. However if this is a script which has to automatically run at startup, this will be a showstopper. So here is some additional information on how to avoid these warnings.

I didn’t tested this in the context of a computer starting up and executing such a script. I just tested this from the Administrator point of view: you run a PowerShell script from a share and you want to avoid that warning. Now what? I will briefly explain Execution Policies and Execution Policy Scopes before actually presenting a solution.

PowerShell Execution Policies

As most of us know by now, PowerShell comes with an execution policy. This policy define which scripts can ran and from which location. By default it’s configured to restricted:

image

In words:

File \\file.setspn.com\scripts\script.ps1 cannot be loaded because the execution of scripts is disabled on this system.
Please see "get-help about_signing" for more details.
At line:1 char:37
+ \\file.setspn.com\scripts\script.ps1 <<<<
    + CategoryInfo          : NotSpecified: (:) [], PSSecurityException
    + FullyQualifiedErrorId : RuntimeException

However we can easily change the policy to one of the following options:

  • Restricted: no scripts
  • AllSigned: only signed scripts
  • RemoteSigned: local [+detected as local intranet] scripts and signed scripts remotely
  • Unrestricted: all scripts but comes with warnings when running from a share
  • Bypass: all scripts, no warnings

If you want a full explanation on each of these policies, check TechNet: about_Execution_Policies

PowerShell Execution Policy Scope

Now if you changed the above policy using the Set-ExecutionPolicy Policy command, you changed it for everyone running PowerShell scripts on the current machine. Did you know there is an alternative? You can actually specify a scope using the –scope parameter! This can come in very handy when you just want to open up PowerShell temporary for the installation of a product. The following scopes exist:

  • Process: affects only the current PowerShell session
  • CurrentUser: affects only the current user
  • LocalMachine: affects all users on the current computer

If you want a full explanation on each of these scopes, check TechNet: about_Execution_Policies There’s also additional info available if you want to control this by using GPO.

Determing The Active PowerShell Execution Policy

Using the Get-ExecutionPolicy command you can get the current policy, however if you add the –list parameter you can see where it’s coming from:

image

image

Now after running Set-ExecutionPolicy –Scope Process Unrestricted

image

Now the advantage is that whenever this PowerShell session is closed, the overall policy remains to what is was before my modification.

Getting Rid Of the Warning

All of the options below will explain how you can get rid of the warning when execution a script from an UNC path. Between [] I’ve added the PowerShell execution policies this is working with. The PowerShell script I’m running is located on \\file.setspn.com\scripts and it only contains one line: “write-host –forefgroundcolor green “script executed successfully””.

Option 1: use a shortname in the path [RemoteSigned/Unrestricted]

image

Why does this work?

image

By default IE7 and up are configured to automatically detect intranet network. This one checkbox actually is equivalent to checking the 3 options below it. Because we use a UNC path with a short name (\\file), it’s detected as an Intranet and hence no warning is displayed. However once you use a FQDN (\\file.setspn.com), this detection does not work. Reference: Enabling "Automatically Detect Intranet Network" on a domain member computer will enable all the three Intranet Options automatically.

Do we like this? I don’t. I hate NetBIOS names. On to the next option

Option 2: Specify Local Intranet Sites [RemoteSigned/Unrestricted]

The security warning only comes up when the script is ran from an untrusted location, we can add the UNC path to the local intranet. We can do this using one of the following formats:

  1. \\*.setspn.com
  2. file://*.setspn.com
  3. \\file.setspn.com
  4. file://file.setspn.com
  5. *.setspn.com
  6. file.setspn.com

The first 4 options will only affect files accessed from UNC paths, however option 5 & 6 will also involve HTTP/HTTPs traffic. Option 1 & 2 are in fact the same. The same goes for option 3 & 4. Whenever you enter an UNC path like \\location it will automatically be converted to file://location by IE. After adding \\*.setspn.com:

image

The script now runs just fine:

image

Option 3: Copy The Script Locally [RemoteSigned/Resctricted]

Obviously this is not applicable in certain scenario’s, but as a quick work around it’s a possibility.

image

Option 4: Sign your scripts [AllSigned/RemoteSigned/Resctricted]

Another possibility is to sign your scripts. A detailed guide is out of the scope of this post, a good description of the process can be found here: Scott Hanselman: Signing PowerShell Scripts

Option 5: Use The Bypass ExecutionPolicy [Resctricted/AllSigned/RemoteSigned/Resctricted]

All of the above options (beside #1) require you to do some modifications. There’s also a way to just suppress these warnings: the Bypass exeuction policy! I think you can apply it in one of the following ways:

  1. Set-ExecutionPolicy Bypass
  2. Set-ExecutionPolicy Bypass –Scope Process
  3. Powershell.exe –ExecutionPolicy Bypass –file \\file.setspn.com\script.ps1

Option 1 just doesn’t feel right. I myself would definitely go for option 2 or 3. They are more or less equivalent. Option 3 is ideally for calling a PowerShell script from within a bat file. Option 2 is just neat as it’s only active temporarily.

Conclusion

I’m not favoring one of the above options. I think PowerShell execution policies should be part of the Workstation/Server design. A given policy should be decided upon and pushed through GPO. Once that baseline is established, I would choose one of the above options to make a certain script work in the given scenario.

Like if RemoteSigned would be active on workstations, I would consider adding the UNC path to the Local Intranet sites. On the other hand if I would be administering servers (RemoteSigned active) and I’d have a script which I have to run just once, I would consider changing the execution policy to Bypass just for this PowerShell.exe instance (-scope process).

Happy PowerShelling!

3 comments

Windows 2008 R2: Accounts: Administrator Account Status Not Working

Published on Monday, April 4, 2011 in ,

One of the things a colleague of mine encountered in the past, and which I stumbled upon lately is the following. Sometimes people want to have the Local Administrator account disabled on their servers. There has been a GPO to do this for ages. It’s located below Computer Settings > Windows Settings > Security Settings > Local Policies > Security Options. The setting is “Accounts: Administrator Account Status”: Disabled.

The screenshot shown below is from the security policy on a server which has the policy (Administrator Status: disabled) applied. You can see that A group policy is setting the setting to enabled. Which is in fact the opposite of what I have configured through the GPO.

image

One could think I have another GPO being applied later. But using gpresult /H:report.html I can clearly see “my” GPO is winning and that the setting in fact should be set to disabled…

image

Also a regular Resultant Set Of Policy shows the setting as disabled…

image

But the account is Active and remains in this state…

image

image

So, Group Policy Preferences to the rescue! It’s not a real answer as to why things are going wrong, but it’s definitely a doable workaround. This policy works flawless.

image

You can’t always get to the bottom of things…

2 comments

RCDC Not recognized After FIM Configuration Migration

Published on Sunday, April 3, 2011 in

I’m not going to be explain something new today, but I’m just writing this article to explain how I established I was having this issue. Above that I want to make sure people find the explanation easier. This article references a PPT which has some great info on the issue.

It all started with a perfectly fine FIM deployment in a lab environment. One of the things we do from time to time is migrate the configuration to the Acceptance environment. To do this we use the FIM Configuration Migration scripts. After one of the migrations our User Edit RCDC in the Acceptance environment was broken. The RCDC defaulted to the Admin view and stated: There is an error in the synchronizationRule display configuration. Please contact your system administrator.

As I recently reviewed a RCDC troubleshooting article I knew what I had to do:

Make sure to change the level from “Error” to “Verbose”. This will give you the following entry in the event log:

image

In words:

The Resource Management Portal detected an error using the Resource Control Display Configuration (RCDC).  This prevented the portal from displaying the object as expected and the portal switched over to Admin View.

The failure is due to a incorrect configuration file.  The file does not validate against the configuration file schema.

Verify that the configuration file is valid XML and matches the configuration file schema. Either upload a new file or modify the existing file in the Resource Management Portal directory.  Afterward, reset IIS.

So now we get directed to an invalid RCDC XML. No worries: Craig Martin to the rescue: RCDC Troubleshooting He explains how you can use Visual Studio to validate the RCDC XML against the MS schema for RCDCs. The RCDC seemed fine. So eventually we logged a PSS case and got pointed to the following MS PPT:

The PPT is a must read. It clearly states how you end up in this situation and what to do to avoid it. For a fix, without having to wipe your database you’ll have to log a PSS case. The PSS people can give you a SQL procedure which can fix the guid of a given RCDC.

Related forums posts:

P.S. Taking a backup from your FIM Service database is a not a luxury when migration configurations. Make sure you have that safety net available!

0 comments

FIM 2010: Language Pack Update 1 Install Issue

Published on Sunday, March 13, 2011 in

Again I’m posting with some FIM 2010 Update 1 issue. I’m not trying to make a statement regarding the stability of the FIM software, I’m just active in an instable environment Sad smile. This error I received when trying to update a FIM 2010 Service and Portal Language Pack installation to Update 1.

The installer for Update 1 is a next next finish, but somewhere in the middle an application error occurs and the rollback is performed.

The error:

image

In words:

Microsoft.ResourceManagement.Setup.LanguagePack.Resource has stopped working.

I had absolutely no clue what the cause could be. I was logged on using a FIM Installer Account, so permissions should have been fine. On the list of todo’s for this environment I also had a WSS security hotfix. This is the hotfix I mentioned in WSS Killer Security Update. I prefer to install it in a controlled way instead of receiving it through WSUS. However installing the update didn’t work out. The setup just failed. The log files pointed me to: KB939308: Error message when you try to modify or to delete an alternate access mapping in Windows SharePoint Services 3.0: "An update conflict has occurred, and you must re-try this action"

After following the actions in that KB article, I reran the installer of the hotfix and everything worked out fine.

And guess what: the Language Pack Update 1 installer finished just fine too! I have no proof that they are related, but I ran the update multiple times, every time resulting in a crash. Once I cleared the cache of WSS as described in the article the update ran fine.

Happy updating!

2 comments

FIM 2010 Update 1 Installation Issue

Published on in

In the release notes of FIM 2010 Update 1 (KB978864), one of the things mentioned is that you have to make sure the Portal is reachable on http://localhost. Another known issue for things to go wrong seems to be the FIM Service Certificate. During the installation you get the following options:

clip_image002

From Microsoft PSS I heard that there’s a known issue to upgrade to FIM 2010 Update 1 if you choose a customized certificate. One of the requirements for the FIM Service certificate is, is that it has CN=ForefrontIdentityManager in it’s subject. My customer had generated a custom certificate from their internal CA, and of course the subject was different from the required one.

This caused the update to fail and rollback. The following errors were shown in the Application event log:

Entries from the event log, first line logged first:

  • Error : MicrosoftILMPortalCommonDlls.wsp already exists
  • An error occurred while deploying FIM portal solution packs.
  • Error : MicrosoftIdentityManagement.wsp already exists
  • An error occurred while deploying FIM portal solution packs.
  • Error : ILMPasswordPortal.wsp already exists
  • An error occurred while deploying FIM portal solution packs.

To resolve this situation you can run the RTM installer again, but now chose “change”. You’ll be prompted to fill in all setup questions again, but now you can choose “Generate a new self-signed certificate”. After running this successfully you can try to update again.

Some other items I found on my quest for a solution:

Installing update failed because sharepoint not installed on "localhost"?

In my opinion removing .WSP’s your self in WSS is not a great idea. The FIM Update installers really depend on the fact that they expect the .WSP’s to be in place. If you start messing with them you might break things completely. If you are having issues reaching your portal at http://localhost, verify the bindings for the SharePoint site in IIS. You could add:

  • 127.0.0.1 80
  • ::1 80

To ensure proper access to http://localhost. In case you don’t have “all available addresses 80” set as the binding.

Another possible solution: ILM 2 Beta 3 Premature Failure - ilmpasswordportal.wsp already exists Again, I would really advise against deleting .WSP’s yourself. Even if they are in error, try running the FIM Service & Portal setup in Change mode. You’ll see it will re-deploy the .WSP solutions.

P.S. If you want detailed information regarding a failure for an update, try running the update.msp file like this: msiexec /p update.msp /L*V c:\update.log

1 comments

FIM MA Full Import Broken

Published on in

One of my customers had the following error when they ran a Full Import on the FIM MA: app-store-import-exception

clip_image002

In the application event log I found the following error which was occurring every time they ran a Full Import.

clip_image002[6]

A copy paste of the error:

The description for Event ID 6500 from source FIMSynchronizationService cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

If the event originated on another computer, the display information had to be saved with the event.

The following information was included with the event:

There is an error executing ILM MA full import.
Type: System.ArgumentNullException

Message: Value cannot be null.
Parameter name: key

Stack Trace:    at System.ThrowHelper.ThrowArgumentNullException(ExceptionArgument argument)
   at System.Collections.Generic.Dictionary`2.FindEntry(TKey key)
   at System.Collections.Generic.Dictionary`2.TryGetValue(TKey key, TValue& value)
   at Microsoft.ResourceManagement.Schema.ServerSchemaManager.GetAttributeSchema(String attributeName)
   at Microsoft.ResourceManagement.Query.QueryProcessor.ReadFragment(SqlDataReader reader, Int64& resultCount, Boolean& endOfSequence)
   at Microsoft.ResourceManagement.Query.QueryProcessor.ReadQueryResults(SqlDataReader reader, Int64& resultCount, Boolean& endOfSequence)
   at Microsoft.ResourceManagement.Query.QueryProcessor.ReadQueryResults(SqlDataReader reader)
   at Microsoft.ResourceManagement.Data.Sync.FullImportGetNext(Int64 beginObjectKey, Int64 maxObjectKey, Int32 batchSize)
   at MIIS.ManagementAgent.RavenMA.FullImportGetNextBatch(Int64 maxObjectKey, Int32 batchSize)

the message resource is present but the message is not found in the string/message table

A search on Google led me to the following TechNet forum post: App-Store-Import-Exception

And then I spent quit some time on SQL tracing and trying to find out which key was null. I found none… I don’t know how I got to it, but I figured refreshing the schema couldn’t hurt to perhaps get a more descriptive error. It seemed that someone had been modifying the schema!:

clip_image002[8]

After the schema refresh imports started running again. It’s obvious to refresh the schema if you changed something to the schema in the Portal. However if you start from the given error it wasn’t as obvious to refresh the schema… Case Solved!